Re: killed ctview objects

[jmi@columbo.inria.fr (Janet Bertot)]

In article <31852@sophia.inria.fr>, Guido.Bosch@loria.fr (Guido Bosch) writes:
=> Is there a "silent way" to test whether a ctview object has been "killed" ?
=> 
=> By "killed", I mean: If one has clicked with the mouse on the little
=> hollow rectangle in the upper left corner, which makes it disappear
=> forever. 

What the little button does is kill all the windows associated with a ctview
by sending a message terminate throughout the widget hierarchy.  Note that 
even "after killing the ctview" the associated lisp structure still exists.  
The silent test is:

	(windowp (send 'window <ctview>)) 

which will be t if the window exists (actually it returns the lisp window
structure), otherwise it is ().

=> By "silent", I mean the following: There is a "noisy" way:
=> 
=> 		(send 'show <ctview>)
=> 
=> which returns the number 1 if the <ctview> has been killed. But if it
=> hasn't, it will be exposed, so this cannot be used as a "silent" test
=> without visible side effects.

Yes, well the message "show" makes the window visible if it exists, otherwise
marks the window to be created visible when it is eventually mapped.  The show
and hide messages allow one to create objects where sub-widgets are not forced
to be visible when mapped.  


	--janet